Intelligent arterial catheter infusion chemotherapy system

Abstract

The present invention discloses an intelligent arterial catheter perfusion chemotherapy system, which relates to the technical field of medical devices. The system includes the following components: an image processing module: collecting CT and MRI medical image data of patients, and using advanced image processing algorithms to process the images to extract three-dimensional models of blood vessels and key tumor structures. The present invention realizes accurate recognition and three-dimensional modeling of blood vessels and tumor structures by deeply integrating medical image processing technology and advanced algorithm models. This process not only improves the topological construction accuracy of the vascular network, but also makes the identification of the tumor location and its surrounding sensitive organs more accurate. Combined with the dual optimization of algorithms and genetic algorithms, the catheter path planning is not only fast and effective, but also fully considers the hemodynamic characteristics, ensuring that the catheter can advance to the tumor supply artery along the optimal and safe path, thereby improving the accuracy of the operation and reducing errors and risks during the operation.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to an intelligent arterial catheter perfusion chemotherapy system. Background Art

[0002] With the rapid development of medical technology, precision medicine has become an important trend in modern medicine. In complex surgeries, especially those involving blood vessels and tumor treatment, accurate identification and positioning of vascular networks and tumor structures are crucial. Traditional surgical methods often rely on the doctor's experience and two-dimensional imaging data, which makes it difficult to fully and intuitively display the three-dimensional structure of blood vessels and tumors, resulting in high errors and risks during the operation. The development of an intelligent arterial catheter perfusion chemotherapy system is a key technical requirement to improve surgical accuracy and reduce surgical risks.

[0003] Traditional arterial catheter perfusion systems can achieve the identification and positioning of blood vessels and tumors to a certain extent, but there are still many shortcomings in path planning. First, the topological construction of the vascular network has limited accuracy, which makes it difficult to accurately reflect the true morphology and complex structure of the blood vessels, resulting in inaccurate path planning. Second, traditional algorithms do not take into account enough when dealing with hemodynamic characteristics, and it is easy to ignore the impact of blood flow on the catheter path, increasing the risk during the operation. In addition, the traditional system lacks real-time navigation and feedback mechanisms, and cannot immediately correct deviations during the operation, limiting the flexibility and efficiency of surgical operations.

[0004] Therefore, in response to the shortcomings of traditional medical navigation technology, the development of an intelligent arterial catheter perfusion chemotherapy system has not only significantly improved the accuracy and safety of the surgery, but also enhanced the real-time and flexibility of the surgery, bringing patients a better treatment experience and faster recovery. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide an intelligent arterial catheter perfusion chemotherapy system. It can achieve optimal planning and safe navigation of the catheter path through precise three-dimensional modeling and advanced algorithm optimization. The real-time tracking and feedback mechanism ensures accurate and efficient surgery, reduces risks, shortens time, brings patients a better treatment experience and rehabilitation effect, and leads precision medicine to a new height.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: an intelligent arterial catheter perfusion chemotherapy system, which includes the following components:

[0007] Image processing module: collects the patient's CT and MRI medical image data, and processes the images using advanced image processing algorithms to extract the three-dimensional models of blood vessels and key tumor structures;

[0008] Navigation planning module: Based on the pre-processed medical imaging data, the system's built-in intelligent navigation and path planning algorithm can automatically plan the optimal catheter path according to the preset target points and avoidance points;

[0009] Real-time navigation module: During the operation, the planned catheter path is superimposed on the real-time image through real-time image tracking and navigation technology, providing doctors with intuitive navigation information. It also has a real-time feedback function, which can monitor the position and status of the catheter and adjust the path planning in time to ensure the smooth progress of the operation;

[0010] Precision perfusion module: After the catheter accurately reaches the tumor-feeding artery, anticancer drugs and embolic agents are injected into the tumor tissue;

[0011] Postoperative optimization module: After the operation is completed, the surgical effect is evaluated by comparing the pre-operative and post-operative medical imaging data. According to the evaluation results and the patient's feedback information, the system automatically adjusts the built-in intelligent navigation and path planning algorithms.

[0012] Furthermore, the image processing module receives DICOM format medical image data generated by CT and MRI scanning equipment, uses a threshold segmentation algorithm to identify and segment regional position images of blood vessels and tumors, stacks and converts the two-dimensional image data into three-dimensional volume data, and generates a three-dimensional model of key structures of blood vessels and tumors based on the three-dimensional volume data.

[0013] Furthermore, the image processing module receives DICOM format medical image data generated by CT and MRI scanning equipment, uses a threshold segmentation algorithm to identify and segment the regional position images of blood vessels and tumors, calculates the grayscale histogram of the image through the threshold segmentation algorithm, understands the grayscale distribution of the image, and sets the threshold value to T based on the grayscale mean. The grayscale mean calculation formula is: ,in and are the minimum and maximum grayscale values, respectively. For each pixel in the image, if its grayscale value is greater than or equal to the threshold T, it is marked as a blood vessel and tumor area, otherwise, it is marked as a background area. The mathematical expression is: ,in, represents the tag value after segmentation, Represents the original image at coordinates After the segmentation is completed, the preliminary blood vessel and tumor areas are obtained, and then after morphological processing, noise and small isolated points are removed to obtain accurate blood vessel and tumor area segmentation results. The two-dimensional image data are stacked and converted into three-dimensional volume data, and a three-dimensional model of the key structures of blood vessels and tumors is generated based on the three-dimensional volume data.

[0014] Furthermore, the navigation planning module receives the three-dimensional blood vessel and tumor model generated by the image processing module, extracts the blood vessel centerline from the three-dimensional model using the centerline extraction algorithm, constructs a blood vessel network topology map based on the blood vessel centerline data, sets the tumor position as the target position and the sensitive organ as the avoidance point in the blood vessel network topology map, and uses the The algorithm conducts a preliminary exploration of the catheter path in the constructed vascular network topology map, and uses a genetic algorithm to optimize the catheter path according to the hemodynamic characteristics, thereby reducing friction between the catheter and the vascular wall and reducing trauma to the patient.

[0015] Furthermore, the navigation planning module uses a centerline extraction algorithm to extract the vascular centerline from the three-dimensional model, constructs the volume data V (x, y, z) of the vascular model of the vascular network topology, and calculates the gradient information of the vascular model. The gradient can be expressed as: By tracing the direction of the gradient, the central axis of the blood vessel is found, the center line extracted initially is refined, and the refined center line nodes are used as vertices V, and the connection relationship of the center line is used as edges E to construct the topological map of the vascular network. , where V is the vertex set, E is the edge set, and the edge weight is calculated based on the length of the blood vessel. and diameter The characteristic data is calculated, and the edge Connect Vertices and , then the weight It is expressed as: , where f is a custom function , optimize the constructed topology graph and remove unreasonable connections and abnormal vertices.

[0016] Furthermore, the navigation planning module utilizes The vascular network topology constructed by the algorithm A preliminary exploration of the catheter path is carried out in the following steps:

[0017] (1) Representing the vascular network as a vascular network topology graph , where V is the vertex set, representing the bifurcation points and key points of the blood vessels, and E is the edge set, representing the connections between vertices;

[0018] (2) Heuristic function Used to estimate the distance or cost from the current vertex n to the target vertex. The coordinates of the current vertex n are , the coordinates of the target vertex G are ( , , ) Use Euclidean distance as a heuristic function to estimate the distance from the current vertex to the target vertex. The heuristic function of Euclidean distance is: ;

[0019] (3) Create two sets: an open set Open and a closed set Closed. The open set Open is used to store vertices to be explored, and the closed set Closed is used to store vertices that have been explored. Add the starting vertex S to the open set Open and set its actual cost , estimated total cost ;

[0020] (4) In each loop, select the estimated total cost from the open set Open The smallest vertex n is removed from the open set Open and added to the closed set Closed. For all adjacent vertices m of vertex n, the new actual cost of reaching vertex m through vertex n is calculated. ,in, is the weight of the edge from vertex n to vertex m. Vertex m is not in the open set Open. Add it to the open set Open and set , estimated total cost , and record that the parent vertex of vertex m is n, vertex m is already in the open set Open, and ,renew , estimated total cost , and update the parent vertex of vertex m to n;

[0021] (5) Loop until the target vertex G is added to the closed set Closed, and then trace back from the target vertex G to the starting vertex S to obtain the preliminary catheter path.

[0022] Furthermore, the navigation planning module uses a genetic algorithm to optimize the catheter path, represents the catheter path as a chromosome code, randomly generates a group of chromosomes as the initial population, the population size is recorded as N, and defines the fitness function F according to the hemodynamic characteristics. Assuming that the blood flow velocity is v, the blood flow pressure is P, and the vascular wall shear stress is , then the fitness function is expressed as: ,in is the weight of each feature, is the corresponding function. The roulette wheel selection method is used to select chromosomes with higher fitness from the current population to generate the parents of the next generation population. The selected parent chromosomes are crossovered with a certain probability to generate new daughter chromosomes. The daughter chromosomes are mutated with a smaller probability. The daughter chromosomes obtained after selection, crossover and mutation operations are combined into a new population. The above operations are repeated until the fitness of the population is no longer significantly improved. The chromosome with the highest fitness in the current population is output as the optimized catheter path.

[0023] Furthermore, during the operation, the real-time navigation module obtains real-time image data of the patient's body through the high-precision image acquisition equipment in the operating room, automatically identifies the characteristics of the catheter using the traceable markers set on the catheter, and accurately tracks the three-dimensional position and direction of the catheter in the patient's body in real time. The optimal catheter path calculated before the operation is superimposed on the real-time image, and the actual position and status of the catheter and related parameters in the patient's body are monitored in real time. The planned path is compared and analyzed, and a feedback mechanism is immediately triggered when deviations and abnormalities are found. The path is manually adjusted to ensure that the catheter moves along the optimal path. An intuitive and easy-to-use user interface is provided, and relevant information is displayed graphically on the display screen.

[0024] Furthermore, the precise perfusion module prepares in advance an appropriate amount of anticancer drugs and embolic agents that have been strictly quality-controlled and proportion-calculated according to the patient's condition and treatment plan, sets the mixing ratio through a preset program, and automatically mixes according to the set ratio and sequence. Before perfusion, the perfusion speed and flow are set using a precise flow control system according to the patient's physiological parameters, tumor characteristics and treatment plan. After confirmation, automatic perfusion is started, and the mixed drugs are injected into the tumor-feeding artery through the catheter. During the perfusion, the patient's vital signs, perfusion parameters and catheter position are monitored in real time. In case of abnormality, an alarm is issued and adjustments are prompted. The perfusion automatically stops after reaching the predetermined perfusion volume or time, and a perfusion report is generated.

[0025] Furthermore, the postoperative optimization module collects preoperative and postoperative medical imaging data, uses image registration technology for precise alignment, compares and detects changes in the surgical area, evaluates the surgical results based on the expert opinions of clinical doctors, and establishes a postoperative feedback mechanism for patients. Based on the evaluation and feedback, the system automatically adjusts and optimizes the internal algorithm model parameters for continuous iterative improvement.

[0026] Compared with the existing technology, the intelligent arterial catheter perfusion chemotherapy system and method have the following beneficial effects:

[0027] 1. This invention realizes the accurate recognition and 3D modeling of blood vessels and tumor structures by deeply integrating medical image processing technology and advanced algorithm models. This process not only improves the topological construction accuracy of the vascular network, but also makes the identification of the tumor location and its surrounding sensitive organs more accurate. The dual optimization of algorithms and genetic algorithms makes catheter path planning not only fast and effective, but also fully considers the hemodynamic characteristics, ensuring that the catheter can advance along the optimal and safe path to the tumor supply artery, improving the accuracy of the operation and reducing errors and risks during the operation.

[0028] 2. The present invention realizes real-time three-dimensional position and direction tracking of the catheter in the patient's body by integrating high-precision image acquisition equipment and traceable marker technology. This function not only provides intuitive visual feedback for the surgeon, but also enables any deviations and abnormalities during the operation to be discovered and corrected immediately. By superimposing the optimal catheter path planned preoperatively on the real-time image, the doctor can clearly compare the difference between the actual path and the planned path, and thus make rapid adjustment decisions.

[0029] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0031] Figure 1 This is the process operation diagram of the intelligent arterial catheter perfusion chemotherapy system;

[0032] Figure 2 Flow chart of the intelligent arterial catheter infusion chemotherapy system. DETAILED DESCRIPTION

[0033] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Embodiment 1:

[0035] This embodiment provides an intelligent arterial catheter perfusion chemotherapy system, which deeply integrates medical image processing technology and advanced algorithm models to achieve three-dimensional accurate recognition and modeling of blood vessels and tumor structures. The dual optimization with genetic algorithms can plan the optimal and safe catheter path to directly reach the tumor supply artery. The real-time navigation module uses high-precision image acquisition and tracking technology to ensure real-time feedback and instant adjustment during the operation, greatly improving the accuracy, safety and flexibility of the operation.

[0036] The image processing module receives the DICOM format image data generated by the CT and MRI scanning equipment, calculates the grayscale histogram of the received image, understands the grayscale distribution of the image, and sets the threshold T based on the grayscale mean. The grayscale mean calculation formula is: ,in and are the minimum and maximum grayscale values, respectively. For each pixel in the image, if its grayscale value is greater than or equal to the threshold T, it is marked as a blood vessel and tumor area, otherwise, it is marked as a background area. The mathematical expression is: ,in, represents the tag value after segmentation, Represents the original image at coordinates The grayscale values ​​at the locations are used to perform morphological processing on the initially segmented blood vessels and tumor regions to remove noise and small isolated points, thereby obtaining more accurate blood vessel and tumor region segmentation results. The two-dimensional image data are stacked into three-dimensional volume data according to the spatial position relationship, and a three-dimensional reconstruction algorithm is used to generate a three-dimensional model of the blood vessels and tumors. The three-dimensional model is smoothed to improve the visual effect and accuracy of the model.

[0037] The navigation planning model receives the generated three-dimensional blood vessel and tumor model, extracts the blood vessel centerline from the three-dimensional blood vessel model using the centerline extraction algorithm, and constructs a blood vessel network topology map based on the blood vessel centerline data. The vertices in the topology map represent the blood vessel bifurcation points and key points, and the edges represent the blood vessel segments. The weights of the edges include the length, curvature, and diameter parameters of the blood vessel segments. The tumor location and sensitive organ location are clearly marked in the topology map, and the tumor location is set as the target location, and the sensitive organ is set as the avoidance point. The algorithm conducts a preliminary exploration of the catheter path in the constructed vascular network topology map and sets The algorithm starts with the estimated catheter insertion point and ends with the tumor location. It selects a suitable heuristic function to estimate the cost from the current vertex to the target vertex and runs The algorithm searches for the preliminary optimal path from the starting point to the end point in the vascular network topology map, optimizes the path using a genetic algorithm, defines an objective function, and considers the length, curvature, friction with the vascular wall, and avoidance of sensitive organs of the catheter path, aiming to minimize the combined impact of these adverse factors. The catheter path is represented as an individual in the genetic algorithm, and each vertex and the connection between vertices on the path are used as genes of the genetic algorithm. Through selection, crossover, and mutation genetic operations, the population is continuously iterated to evolve and find a better catheter path. In each iteration, the objective function is used to evaluate the fitness of each individual, and individuals with high fitness are selected to enter the next generation. When the fitness improvement is no longer significant, the evolution process is stopped, and the optimal catheter path optimized by the genetic algorithm is output. The optimized path is verified based on the hemodynamic characteristics and the experience of clinicians to ensure its feasibility and safety in actual operation.

[0038] The real-time navigation module obtains real-time imaging data of the patient's body through high-precision imaging acquisition equipment in the operating room, pre-installs tiny metal markers that can be identified by imaging equipment on the arterial catheter, receives real-time imaging data from the imaging acquisition equipment, and automatically calculates and tracks the three-dimensional position and direction of the catheter by identifying the markers in the image. The optimal path planned before surgery is superimposed on the real-time image through coordinate conversion and mapping technology, so as to achieve real-time synchronization of the virtual path and the real image, and monitor the deviation between the actual position of the catheter and the planned path in real time. Once an abnormality is found or the deviation exceeds the preset threshold, an alarm is triggered immediately, and adjustment suggestions are provided through the user interface. An intuitive and easy-to-use user interface is provided to display real-time images, catheter positions, planned paths, deviation information and any necessary warning information, display catheter paths, target areas and surrounding vascular structures, and provide doctors with operable interface elements for manual adjustments when necessary.

[0039] The precision perfusion module prepares appropriate amounts of anticancer drugs and embolic agents that have undergone strict quality control and ratio calculations according to the patient's specific condition and treatment plan. The mixing ratio of anticancer drugs and embolic agents is set according to treatment needs. This ratio is based on the doctor's prescription and the patient's physiological characteristics to ensure that the drugs and embolic agents can achieve the best therapeutic effect after mixing. Before the perfusion begins, the intelligent drug mixing system in the precision perfusion module is used to automatically mix the anticancer drugs and embolic agents in the set ratio and sequence. According to the patient's physiological parameters, tumor characteristics and preset treatment plan, the appropriate perfusion speed is set through the system's precise flow control system. and flow rate to ensure that drugs and embolic agents can be injected into tumor tissue in the best way while reducing damage to the patient's normal tissues. After confirming that all parameter settings are correct and the catheter is in the right position, automatic perfusion is performed. The real-time monitoring function continuously monitors the patient's vital signs, perfusion parameters, and key information on the catheter position. Once any abnormality or deviation is found, an alarm is immediately issued, prompting the doctor to make necessary adjustments to ensure the accuracy and safety of the perfusion. When the predetermined perfusion volume or treatment time is reached, the system automatically stops perfusion and generates a detailed perfusion report. To ensure the safety of the patient, the precision perfusion module is also equipped with multiple safety protection mechanisms.

[0040] Postoperative optimization module: Before the operation, high-resolution medical imaging equipment is used to conduct a comprehensive scan of the patient to obtain detailed cardiac structure images. The collected image data is imported into the medical image processing module of the system for preliminary image preprocessing. According to the same scanning protocol and parameters, medical image data of the patient is collected again, and the postoperative image data is also imported into the system and preprocessed accordingly. The preoperative and postoperative medical image data are accurately aligned, and the changes in the surgical area before and after the operation are automatically compared. A team of experts in related fields is organized to review and evaluate the changes in the surgical area automatically detected by the system. A postoperative feedback questionnaire is designed to cover the patient's subjective feelings, symptom improvement, and changes in quality of life. The evaluation results of clinical experts are combined with the patient's subjective feedback to form a comprehensive evaluation report on the effect. The algorithm model's performance in image registration and change detection is automatically analyzed, and the optimized algorithm model is applied to new patient cases, and medical imaging data and patient feedback continue to be collected.

[0041] Embodiment 2:

[0042] Based on the first embodiment, the present invention particularly emphasizes the navigation planning module model;

[0043] First, a three-dimensional blood vessel and tumor model generated by an image processing module is received.

[0044] Secondly, the centerline extraction algorithm is used to extract the vascular centerline data from the 3D model, and the vascular network topology is constructed based on the vascular centerline data. The gradient information of the vascular model is calculated based on the volume data V volume (x, y, z) of the vascular model. The gradient can be expressed as: By tracing the direction of the gradient, the central axis direction of the blood vessel is found, the initially extracted centerline is refined, and the refined centerline nodes are used as vertices V, and the connection relationship of the centerline is used as edges E to construct the vascular network topology map , where V is the vertex set, E is the edge set, and the edge weight is calculated based on the length of the blood vessel. and diameter The characteristic data is calculated, and the edge Connect Vertices and , then the weight It is expressed as: , where f is a custom function , optimize the constructed topology graph and remove unreasonable connections and abnormal vertices.

[0045] Then, the vascular network is represented as a vascular network topology graph , where V is the vertex set, E is the edge set, and the Euclidean distance is used as the heuristic function to estimate the distance from the current vertex to the target vertex. The heuristic function of the Euclidean distance is: , initialize the open set (Open) and the closed set (Closed), add the starting vertex to the Open set, and set its actual cost and estimated total cost, loop the following steps until the target vertex is added to the Closed set, select the vertex with the smallest estimated total cost from the Open set, remove it from the Open set and add it to the Closed set, calculate the new actual cost and estimated total cost for all adjacent vertices of the vertex, and update the Open set, trace back from the target vertex to the starting vertex, and obtain the preliminary catheter path.

[0046] Next, the catheter path is represented as a chromosome code, each gene represents the connection between vertices on the path, and a set of chromosomes is randomly generated as the initial population. The population size is recorded as N. The fitness function F is defined according to the hemodynamic characteristics. Assume that the blood flow velocity is v, the blood flow pressure is p, and the vascular wall shear stress is , then the fitness function is expressed as: ,in is the weight of each feature, is the corresponding function. The roulette wheel selection method is used to select chromosomes with higher fitness from the current population as parents. The selected parent chromosomes are crossovered with a certain probability to generate new daughter chromosomes. The daughter chromosomes are mutated with a smaller probability to increase the diversity of the population. The selection, crossover, and mutation operations are repeated until the fitness of the population is no longer significantly improved and the preset number of iterations is reached. The chromosome with the highest fitness in the current population is output, which is the optimized catheter path.

[0047] Finally, based on the experience of clinicians and the results of hemodynamic simulation, the optimized catheter path is verified to ensure its feasibility and safety in actual operation, and necessary adjustments are made to the path based on the verification results to meet actual treatment needs.

[0048] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. Intelligent arterial catheter perfusion chemotherapy system, characterized by: The system consists of the following components: Image processing module: collects the patient's CT and MRI medical image data, processes the images using image processing algorithms, and extracts the three-dimensional models of key structures of blood vessels and tumors; Navigation planning module: Based on the pre-processed medical image data, the system's built-in intelligent navigation and path planning algorithm automatically plans the optimal catheter path according to the preset target points and avoidance points, receives the three-dimensional blood vessel and tumor model generated by the image processing module, uses the centerline extraction algorithm to extract the blood vessel centerline from the three-dimensional model, and constructs a blood vessel network topology map based on the blood vessel centerline data. In the blood vessel network topology map, the tumor position is set as the target position, and the sensitive organs are set as avoidance points. The algorithm conducts a preliminary exploration of the catheter path in the constructed vascular network topology map, and uses a genetic algorithm to optimize the catheter path according to the hemodynamic characteristics, thereby reducing friction between the catheter and the vascular wall and reducing trauma to the patient. Real-time navigation module: During the operation, the image data of the patient's body is obtained in real time through the image acquisition equipment in the operating room. The characteristics of the catheter are automatically identified by the traceable markers set on the catheter. The three-dimensional position and direction of the catheter in the patient's body are tracked in real time. The optimal catheter path calculated before the operation is superimposed on the real-time image through coordinate conversion and mapping technology. The actual position, state and related parameters of the catheter in the patient's body are monitored in real time, and compared and analyzed with the planned optimal catheter path. When deviations and abnormalities are found, the feedback mechanism is immediately triggered, and the path is adjusted manually to ensure that the catheter moves along the optimal path. An intuitive and easy-to-use user interface is provided, and relevant information is displayed graphically on the display screen; Precision perfusion module: According to the patient's condition and treatment plan, anticancer drugs and embolic agents that have been quality controlled and calculated are prepared in advance, the mixing ratio is set through a preset program, and the intelligent drug mixing system is used to automatically mix according to the set ratio and sequence. Before perfusion, the perfusion speed and flow rate are set according to the patient's physiological parameters, tumor characteristics and treatment plan using a precision flow control system, and the mixed drugs are injected into the tumor-feeding artery through a catheter. During perfusion, the patient's vital signs, perfusion parameters and catheter position are monitored in real time. In case of abnormality, an alarm is issued and adjustments are prompted. After reaching the predetermined perfusion volume or time, the perfusion stops automatically and a perfusion report is generated; Postoperative optimization module: After the operation is completed, the preoperative and postoperative medical imaging data are collected, and the image registration technology is used to accurately align them. The changes in the surgical area are compared and detected. The surgical effect is evaluated in combination with the expert opinions of clinical doctors. A postoperative feedback mechanism for patients is established. Based on the evaluation and feedback, the system automatically adjusts the parameters of the intelligent navigation and path planning algorithms.

2. The intelligent arterial catheter perfusion chemotherapy system according to claim 1, characterized in that: The image processing module receives DICOM format medical image data generated by CT and MRI scanning equipment, uses the threshold segmentation algorithm to identify and segment the regional position images of blood vessels and tumors, calculates the grayscale histogram of the image through the threshold segmentation algorithm, understands the grayscale distribution of the image, and sets the threshold value to T based on the grayscale mean. The grayscale mean calculation formula is: ,in and are the minimum and maximum grayscale values, respectively. For each pixel in the image, if its grayscale value is greater than or equal to the threshold T, it is marked as a blood vessel and tumor area, otherwise, it is marked as a background area. The mathematical expression is: ,in, represents the tag value after segmentation, Represents the original image at coordinates After the segmentation is completed, the preliminary blood vessel and tumor areas are obtained, and then after morphological processing, noise and small isolated points are removed to obtain accurate blood vessel and tumor area segmentation results. The two-dimensional image data are stacked and converted into three-dimensional volume data, and a three-dimensional model of the key structures of blood vessels and tumors is generated based on the three-dimensional volume data.

3. The intelligent arterial catheter perfusion chemotherapy system according to claim 1, characterized in that: The navigation planning module uses the centerline extraction algorithm to extract the blood vessel centerline from the three-dimensional model, constructs the volume data V (x, y, z) of the blood vessel model of the blood vessel network topology, and calculates the gradient information of the blood vessel model. The gradient is expressed as: By tracing the direction of the gradient, the central axis of the blood vessel is found, the initially extracted centerline is refined, and the refined centerline nodes are used as vertices and the connection relationship of the centerline is used as the edge E to construct the topological map of the vascular network. , where V is the vertex set and E is the edge set. The edge weight is calculated based on the length and diameter of the blood vessel. Connect Vertices and , the length is , the diameter is , then the weight It is expressed as: , where f is a custom function , optimize the constructed topology graph and remove unreasonable connections and abnormal vertices.

4. The intelligent arterial catheter perfusion chemotherapy system according to claim 1, characterized in that: The navigation planning module utilizes The vascular network topology constructed by the algorithm A preliminary exploration of the catheter path is carried out in the following steps: (1) Representing the vascular network as a vascular network topology graph , where V is the vertex set, representing the bifurcation points and key points of the blood vessels, and E is the edge set, representing the connections between vertices; (2) Heuristic function Used to estimate the distance or cost from the current vertex n to the target vertex. The coordinates of the current vertex n are , the coordinates of the target vertex G are ( , , ) Use Euclidean distance as a heuristic function to estimate the distance from the current vertex to the target vertex. The heuristic function of Euclidean distance is: ; (3) Create two sets: an open set Open and a closed set Closed. The open set Open is used to store vertices to be explored, and the closed set Closed is used to store vertices that have been explored. Add the starting vertex S to the open set Open and set its actual cost , estimated total cost ; (4) In each loop, select the estimated total cost from the open set Open The smallest vertex n is removed from the open set Open and added to the closed set Closed. For all adjacent vertices m of vertex n, the new actual cost of reaching vertex m through vertex n is calculated. ,in, is the weight of the edge from vertex n to vertex m. Vertex m is not in the open set Open. Add it to the open set Open and set , estimated total cost , and record that the parent vertex of vertex m is n, vertex m is already in the open set Open, and ,renew , estimated total cost , and update the parent vertex of vertex m to n; (5) Loop until the target vertex G is added to the closed set Closed, and then trace back from the target vertex G to the starting vertex S to obtain the preliminary catheter path.

Images